A machine learning framework for space medicine predictive diagnostics with physiological signals

Prognostics and health management (PHM) in the context of space missions focuses on the fundamental issues of system failures in an attempt to predict when the failures may occur, and links these issues to system life cycle management. Space missions that are targeting for aerospace exploration or aviation also pose great challenges on the health conditions of people involved, such like astronauts, crew members, aviators, etc. Considering the inherent risks of space missions and the difficulty of direct communications between crew and ground support medical specialists, we see that greater autonomy in medical operations for crew is required. Namely, there is an urgent call for an effective onboard medical system to predict and prevent health problems in a timely manner, rather than following reactive approaches which are inherent to conventional medicine.